François Chung, Ph.D.

Tag: modelización de la apariencia

CVIU 2013 - Artículo de revista científica

CVIU 2013 – Artículo de revista científica

Publicación

François Chung, Hervé Delingette; Regional appearance modeling based on the clustering of intensity profiles; In: Computer Vision and Image Understanding (CVIU), 117 (6), pp. 705-717, 2013.

Abstract

Model-based image segmentation is a popular approach for the segmentation of anatomical structures from medical images because it includes prior knowledge about the shape and appearance of structures of interest. This paper focuses on the formulation of a novel appearance prior that can cope with large variability between subjects, for instance due to the presence of pathologies. Instead of relying on Principal Component Analysis (PCA) such as in Statistical Appearance Models (SAMs), our approach relies on a multimodal intensity profile atlas from which a point may be assigned to several profile modes consisting of a mean profile and its covariance matrix. These profile modes are first estimated without any intra-subject registration through a boosted Expectation-Maximization (EM) classification based on spectral clustering. Then, they are projected on a reference mesh whose role is to store the appearance information in a common geometric representation. We show that this prior leads to better performance than the classical monomodal PCA approach while relying on fewer profile modes.

Palabras clave

  • appearance modeling
  • medical imaging
  • model-based image segmentation
  • unsupervised clustering

Referencias

Publicación

Artículos relacionados

3D Anatomical Human (proyecto INRIA)
Ph.D. Thesis 2011 (tesis doctoral)

LAP 2011 - Libro

LAP 2011 – Libro

Publicación

François Chung; Regional appearance modeling for model-based image segmentation: Methodological approaches to improve the accuracy of model-based image segmentation; Lambert Academic Publishing (LAP), Saarbrücken, 2011; ISBN: 978-3844322095.

Abstract

This thesis presents a novel appearance prior for model-based image segmentation. This appearance prior, denoted as Multimodal Prior Appearance Model (MPAM), is built upon an Expectation–Maximization (EM) clustering of intensity profiles with model order selection to automatically select the number of profile classes. Unlike classical approaches based on Principal Component Analysis (PCA), the clustering is considered as regional because intensity profiles are classified for each mesh and not for each vertex. Comparative results on liver profiles from Computed Tomography (CT) images show that MPAM outperforms PCA-based appearance models. Finally, methods for the analysis of lower limb structures from Magnetic Resonance (MR) images are presented. A first part deals with the creation of subject-specific models for kinematic simulations of the lower limbs. In a second part, the performance of statistical models is compared in the context of lower limb bone segmentation when only a small number of datasets is available for training.

Referencias

Publicación

Libro (Amazon)
Libro (MoreBooks)
Referencia bibliográfica (BibTeX)

Artículos relacionados

3D Anatomical Human (proyecto INRIA)
Ph.D. Thesis 2011 (tesis doctoral)

Más información

LAP – Lambert Academic Publishing

Ph.D. Thesis 2011 - Tesis doctoral

Ph.D. Thesis 2011 – Tesis doctoral

Publicación

François Chung; Regional appearance modeling for deformable model-based image segmentation; Tesis doctoral (Ph.D. Thesis), Mines ParisTech, Centre de Mathématiques Appliquées, 2011.

Abstract

This thesis presents a novel appearance prior for model-based image segmentation. This appearance prior, denoted as Multimodal Prior Appearance Model (MPAM), is built upon an Expectation-Maximization (EM) clustering of intensity profiles with model order selection to automatically select the number of profile classes. Unlike classical approaches based on Principal Component Analysis (PCA), the clustering is considered as regional because intensity profiles are classified for each mesh and not for each vertex.

First, we explain how to build a MPAM from a training set of meshes and images. The clustering of intensity profiles and the determination of the number of appearance regions by a novel model order selection criterion are explained. A spatial regularization approach to spatially smooth the clustering of profiles is presented and the projection of the appearance information from each dataset on a reference mesh is described.

Second, we present a boosted clustering based on spectral clustering, which optimizes the clustering of profiles for segmentation purposes. The representation of the similarity between data points in the spectral space is explained. Comparative results on liver profiles from Computed Tomography (CT) images show that our approach outperforms PCA-based appearance models.

Finally, we present methods for the analysis of lower limb structures from Magnetic Resonance (MR) images. In a first part, our technique to create subject-specific models for kinematic simulations of lower limbs is described. In a second part, the performance of statistical models is compared in the context of lower limb bones segmentation when only a small number of datasets is available for training.

Palabras clave

  • appearance modeling
  • liver
  • lower limbs
  • medical imaging
  • model-based image segmentation
  • unsupervised clustering

Referencias

Publicación

Artículos relacionados

3D Anatomical Human (proyecto INRIA)
CVIU 2013 (artículo de revista científica)
LAP 2011 (libro)

Más información

miccai-2009-acta-de-conferencia-es

MICCAI 2009 – Acta de conferencia

Publicación

François Chung, Hervé Delingette; Multimodal prior appearance models based on regional clustering of intensity profiles; MICCAI 2009: Medical Image Computing and Computer-Assisted Intervention, London, 2009.

Abstract

Model-based image segmentation requires prior information about the appearance of a structure in the image. Instead of relying on Principal Component Analysis (PCA) such as in Statistical Appearance Models (SAMs), we propose a method based on a regional clustering of intensity profiles that does not rely on an accurate pointwise registration. Our method is built upon the Expectation-Maximization (EM) algorithm with regularized covariance matrices and includes spatial regularization. The number of appearance regions is determined by a novel model order selection criterion. The prior is described on a reference mesh where each vertex has a probability to belong to several intensity profile classes.

Referencias

Publicación

Artículos relacionados

Modelo multimodal (proyecto INRIA)
3D Anatomical Human (proyecto INRIA)

Más información

MICCAI – Medical Image Computing and Computer Assisted Intervention

Modelo multimodal

Modelo multimodal

Proyecto INRIA @Sophia-Antipolis, Francia (2009). La segmentación de imágenes basada en modelos requiere información previa acerca de la apariencia de una estructura en la imagen. En lugar de basarse en el análisis de componentes principales (ACP), como se hace con los Statistical Appearance Models (SAM), proponemos un nuevo modelo de apariencia basado en una agrupación regional de perfiles de intensidad que no se basa en un registro punto a punto preciso.

Este modelo multimodal de apariencia, denominado Multimodal Prior Appearance Model (MPAM), se basa en el algoritmo de esperanza-maximización (EM) con matrices de covarianza regularizadas e incluye una regularización espacial. El número de regiones de apariencia esta determinado por un nuevo criterio de selección del orden del modelo. El modelo se describe en una malla de referencia en la cual cada vértice tiene una probabilidad de pertenecer a varias clases de perfil de intensidad.

Probamos nuestro método con 7 mallas del hígado segmentado a partir de imágenes tomográficas (TC) y 4 mallas del tibia segmentado a partir de imágenes de imágenes por resonancia magnética (IRM). Para ambas estructuras, perfiles de intensidad compuestos por 10 muestras extraídas cada mm fueron generados a partir de mallas con alrededor de 4000 vértices.

La principal ventaja de nuestro método es que las regiones de apariencia son extraídas sin necesidad de un registro punto a punto preciso. Otra ventaja es que un modelo puede estar construido con pocos conjuntos de datos (de hecho un conjunto es suficiente). Además, el modelo es multimodal, y por lo tanto capaz de hacer frente a grandes variaciones de apariencia.

Referencias

Artículos relacionados

MICCAI 2009 (acta de conferencia)
Ph.D. Thesis 2011 (tesis doctoral)